| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1998;66:1329-1335
© 1998 The Society of Thoracic Surgeons
a Division of Cardiothoracic Surgery, Department of Surgery, Medical College of Virginia, Richmond, Virginia, USA
b Section of Cardiothoracic Surgery, The Milton S. Hershey Medical Center, Pennsylvania State-Geisinger Health System, Hershey, Pennsylvania, USA
Accepted for publication May 14, 1998.
Address reprint requests to Dr Damiano, Section of Cardiothoracic Surgery, The Milton S. Hershey Medical Center, Pennsylvania State-Geisinger Health System, PO Box 850, Hershey, PA 17033
e-mail: (damiano{at}surg.hmc.psghs.edu)
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Methods. Using the blood-perfused Langendorff technique, isolated rabbit hearts were arrested for 30 minutes of global normothermic ischemia. Cardioplegia consisted of either Krebs-Henseleit solution alone (control) or with pinacidil (50 µmol/L), adenosine (200 µmol/L to 1 mmol/L), or pinacidil + adenosine (200 µmol/L). Recovery of developed pressure and coronary flow were recorded.
Results. Postischemic functional recovery for control, pinacidil, adenosine, and adenosine + pinacidil groups was 44.1% ± 3.4%, 59.5% ± 5.2% (p < 0.05 versus control), 37.0% ± 4.5%, and 56.0% ± 2.9%, respectively.
Conclusions. Adenosine, alone or as adjunct to pinacidil cardioplegia, was not an effective cardioplegic agent, despite shorter times to electromechanical arrest than control. The ineffectiveness of adenosine suggests that the cardioprotective properties of potassium-channel openers involve mechanisms other than the avoidance of membrane depolarization.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Myocardial adenosine triphosphate (ATP)-sensitive potassium (KATP) channels open during situations that result in a decrease in intracellular ATP. This unique class of metabolically linked potassium channels possesses intrinsic cardioprotective properties, opening during cellular ischemia to hyperpolarize the membrane, shorten the action potential, and curtail calcium influx [12]. The resultant contractile failure constitutes an intrinsic energy-sparing mechanism during ischemia. Various drugs have been identified that open these channels at normal intracellular levels of ATP. Hyperpolarized arrest with the KATP channel openers (PCOs) aprikalim, pinacidil, and nicorandil have been studied extensively in our laboratory. These PCOs have been shown to be effective cardioplegic agents that result in reversible electromechanical arrest and adequately protect the myocardium during global ischemia [8–10].
Previous work in our laboratory has demonstrated the cardioprotective properties of the PCOs aprikalim and pinacidil. However, both of these agents had significant shortcomings. They exhibited toxicity at high doses and had narrow therapeutic windows. They were arrhythmogenic and increased postischemic myocardial oxygen consumption [9, 10, 13].
The mechanism of cardioprotection with PCOs is still controversial. It has been hypothesized that action potential shortening and vasodilation may play a role; however, recent studies have shown that neither of these are necessary for PCO-mediated cardioprotection. The avoidance of membrane depolarization that results in the activation of various energy-consuming transport systems to restore ionic equilibrium (the resting membrane potential) also may be an explanation. We hypothesized that myocardial protection would be possible with another known cardioprotective, nondepolarizing agent, adenosine, and that the adjunctive administration of this drug may allow PCO dosage to be minimized and possibly reduce the toxicity of these agents.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Experimental preparation
Preparation of the support animal
The support animal was anesthetized intramuscularly with acepromazine (1 mg/kg) and xylazine (17.5 mg/kg), followed by ketamine (62.5 mg/kg). Anesthesia was monitored throughout the experiment and supplemented as needed. A tracheostomy was performed, an endotracheal tube inserted, and mechanical ventilation begun with 100% oxygen (ventilator model 683; Harvard Apparatus, Dover, MA). Ventilator settings were adjusted to maintain arterial pH between 7.35 and 7.5, carbon dioxide tension between 35 and 45 mm Hg, and oxygen tension greater than 200 mm Hg.
Heparin (2500 U) was given through an ear vein. The right femoral artery was cannulated, the cannula connected to a pressure transducer (model P231D; Gould Inc, Cleveland, OH) and blood pressure continuously monitored on our recording system (Gould ES 1000; Gould Inc). A systolic blood pressure in excess of 80 mm Hg was maintained by transfusion of either blood collected from the donor animal or Plasma-Lyte (Baxter Healthcare Corp, Deerfield, IL). Serial hematocrits were measured.
The left internal jugular vein and the left carotid artery were cannulated. Cannulas were attached to silicone elastomer tubing (internal diameter = 0.125 inch; Baxter Scientific Products, McGaw Park, IL) that was positioned in roller pumps. Arterial blood was pumped (Masterflex model 7013; Cole-Parmer Instrument Co, Chicago, IL) to perfuse a modified Langendorff apparatus described previously [10]. Column height was 80 cm. Effluent from the column was returned to the internal jugular vein of the support animal (Travenol pump model 5M1155; Travenol Laboratories, Inc, Deerfield, IL). Indomethacin (1 mg/kg) was administered to the support animal to augment blood pressure stability. Indomethacin has been shown to have no direct effect on the left ventricular oxygen consumption–systolic pressure–volume area relationship or coronary vascular responsiveness of the isolated heart [9].
Preparation of the donor animal and isolated heart
The donor animal was anesthetized, intubated, ventilated, and heparinized as described above. A rapid cardiectomy was accomplished through a median sternotomy. Blood for transfusion was collected from the thoracic cavity. The heart was placed briefly into a bowl of cold Plasma-Lyte, and a vent (polyethylene tubing, length 2 to 2.5 cm, internal diameter 0.86 mm, Clay Adams, Parsippany, NJ) was inserted into the left ventricle (LV). The aorta was cannulated with the heart suspended from a modified Langendorff apparatus and blood perfusion was begun. A fluid-filled latex balloon was placed into the LV and secured with a pursestring suture in the mitral valve annulus. The balloon was connected by polyethylene tubing to a pressure transducer (model P231D; Gould) and amplifier (model 13-4615-50; Gould). The zero-pressure reference was set at the level of the aortic valve. Two needle electrodes were secured in the right atrial appendage and connected to a pacemaker (model 5320; Medtronic, Inc, Minneapolis, MN). For the duration of the study, heart rate was maintained at a constant rate of 180 to 240 beats/min. Two additional electrodes were placed on the LV epicardium to monitor the bipolar ventricular electrogram. The electrodes were connected to an isolated preamplifier (model 11-G5407-58, Gould Inc) and a universal amplifier (model 13-4615-58, Gould Inc) and filtered between 0.05 and 1,000 Hz. The pressure and electrogram waveforms were displayed continuously and digitized in real time using an AT-CODAS system (DATAQ Instruments, Akron, OH) at a sampling rate of 1,000 Hz.
Coronary flow was measured using an in-line flow probe and continuously monitored using a flow meter (model T206; Transonic Systems, Inc, Ithaca, NY).
The heart was enclosed in a water-jacketed beaker, and myocardial temperature was monitored with a probe placed in the right ventricle (temperature probe model 0112; Shiley Inc, Irvine, CA). Myocardial temperature was maintained at 37°C by adjusting the temperature of the water bath (model 71; Polyscience, Niles, IL). At hourly intervals, heparin (500 U) was administered to the support animal.
Experimental protocol
Hearts that did not generate a systolic pressure exceeding 80 mm Hg at an end-diastolic pressure of 10 mm Hg were excluded from the study. After instrumentation, hearts were given 30 minutes to equilibrate, and baseline data were acquired. Intracavitary LV pressure waveforms and LV bipolar electrograms were recorded at seven balloon volumes, each corresponding to a fixed, intracavitary end-diastolic pressure (EDP; 0, 2.5, 5, 10, 15, 20, and 25 mm Hg). Hearts were randomized to receive 50 mL of one of four cardioplegic solutions before the onset of 30 minutes of global, normothermic ischemia: (1) no cardioplegia (Krebs-Henseleit solution only, control, n = 7); (2) hyperpolarizing cardioplegia using pinacidil (50 µmol/L, n = 8), (3) adenosine (200 µmol/L, n = 8), or (4) pinacidil and adenosine in combination (n = 8). The delivery medium for all cardioplegic solutions was Krebs-Henseleit solution (in mmol/L: NaCl, 118.5; NaHCO3, 25; KCl, 3.2; MgSO4, 1.2; KH2PO4, 1.2; CaCl2, 2.5; and glucose, 5.5). The dose of pinacidil used in this study was based on previous studies in our laboratory that documented the dose–response relationship of pinacidil in this model [9]. The dose of adenosine used was based on other investigations that have demonstrated that a 200 µmol/L dose was cardioprotective in the isolated rabbit heart Langendorff model [14, 15]. In our laboratory, preliminary studies using adenosine in doses up to 1 mmol/L revealed 200 µmol/L to offer optimal functional recovery.
After 30 minutes of global, normothermic ischemia, hearts were reperfused for 30 minutes. The heart was defibrillated if necessary (model D84, Electrodyne Co., Inc., Westwood, MA). Intracavitary LV pressure waveforms and electrograms were recorded over the identical range of balloon volumes recorded during baseline data acquisition. At the conclusion of the study, a sample of the LV was excised, blotted, weighed, and then dried until a constant dry weight was reached. Myocardial edema was expressed as the percent tissue water in the following equation: 
.
Data analysis
Digitized pressure waveforms were collected in data files and analyzed using software developed in our laboratory. Systolic function was determined by the recovery of developed pressure (end-systolic pressure [ESP] minus end-diastolic pressure [EDP]) over a range of LV volumes. The diastolic properties of the LV were measured by comparing mean preischemic LVEDP to postreperfusion LVEDP at a fixed volume.
End-systolic pressure
The ESP of a beat was defined as the maximal point of the digitized pressure waveform. The ESP values of 10 consecutive beats determined the mean ESP. Mean ESP was calculated for each of the seven baseline and seven postreperfusion balloon volumes (V). The ESP versus V data were fitted to a linear ESP–V relationship (ESPVR) with a least-squares linear regression: 
, where Emax is the slope of the ESPVR and k is the y-axis intercept of the ESPVR.
End-diastolic pressure
The EDP of a beat was defined as the point at which the slope of the pressure waveform exceeded 0.5 mm Hg/ms. The EDP for 10 consecutive beats determined the mean ESP and was calculated for each baseline and postreperfusion balloon V. The EDP versus V data were fitted to a linear EDP–V relationship (EDPVR) with a least-squares linear regression: 
, where m is the slope of the EDPVR and V0 is the balloon V at which EDP is zero, or the x-axis intercept. A linear representation of the EDPVR has been shown to be appropriate over the range of V examined in this model [9]. Changes in LVEDP at a single fixed V corresponding to a preischemia LVEDP of 2.5 mm Hg were compared.
Developed pressure
Developed pressure (DP) in the LV was defined as the difference between ESP and EDP for a given beat. The DP of 10 beats was averaged for each balloon V. The DP versus V data were fitted to a linear DP–V relationship using the following regression: 
.
Recovery of developed pressure
The recovery of DP, expressed as a percentage, was calculated as the ratio of the mean postreperfusion DP to the mean baseline DP at the same balloon V. The average percent recovery of DP (%DP) was determined from the following definite integral, approximated by the trapezoidal rule [10]:
 |
Statistical analysis
Results are expressed as the mean ± standard error of the mean. Analysis of variance was used for multiple comparisons. When appropriate, the Kruskal-Wallis analysis of variance on ranks was used as a nonparametric alternative. Individual comparisons between groups were made using a Student-Newman-Keuls post hoc test. A Student’s t test was used for comparisons between two sets, with a Mann-Whitney rank sum test as a nonparametric alternative. A Fisher’s exact test was used to compare mutually exclusive data when appropriate. Differences were considered statistically significant at p less than 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Temporal aspects of the development of electromechanical arrest
Electrical arrest in adenosine, pinacidil, and adenosine + pinacidil groups occurred more rapidly than in controls (Table 1). In addition, hearts treated with pinacidil alone exhibited a more rapid electrical arrest than those treated with adenosine alone. No benefit was observed with the addition of adenosine to pinacidil. Mechanical arrest occurred more rapidly in both the pinacidil and the pinacidil + adenosine groups, compared with the control and adenosine-only groups (see Table 1). There were no statistical differences between pinacidil and pinacidil + adenosine groups.
|
Postischemic diastolic properties
Changes in diastolic compliance were measured by comparing mean preischemic LVEDP to postreperfusion LVEDP at a fixed V (Fig 1). Significant postreperfusion increases in EDPs were observed in the control and adenosine groups (23.4 ± 8.1 mm Hg, p = 0.049; 37.2 ± 13.4 mm Hg, p = 0.036, respectively). However, no significant changes in diastolic compliance were observed in the pinacidil and pinacidil + adenosine groups.
|
|
Coronary flow
Compared with the adenosine and control cardioplegia groups, pinacidil produced a prolonged hyperemic state on reperfusion (Fig 3). Adenosine + pinacidil hearts also experienced a prolonged period of hyperemia compared with those treated with adenosine alone.
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Adenosine, a cardioprotective purine nucleoside, is a receptor-mediated KATP channel agonist. It has demonstrated infarct size reduction in the intact rabbit heart, when used either as a preconditioning agent or an A1 receptor agonist during ischemia [18]. Numerous studies have demonstrated that pretreatment with adenosine provides cardioprotection during ischemia, and that this effect is mediated by A1 receptors [18–22]. Adenosine also has been shown to improve myocardial recovery as an adjunct to hyperkalemic cardioplegia [14, 23]. In an older study, the protective effects of adenosine cardioplegia rivaled those of traditional hyperkalemic solutions [11].
Adenosine also is known to hyperpolarize the myocyte membrane [24]. Moreover, the A1 receptor has been shown to be coupled by a G-protein to the KATP channel. This could potentially augment the cardioprotection seen with PCO cardioplegia alone.
Previous studies using isolated perfused rabbit hearts have shown that adenosine (200 µmol/L) augmentation of hyperkalemic cardioplegia resulted in improved tolerance to ischemia [14]. Dose–response studies (200 µmol/L to 1 mmol/L) in our laboratory demonstrated optimal myocardial recovery with 200 µmol/L adenosine.
Cardioprotective effects of pinacidil versus adenosine
Systolic function and diastolic properties of the LV (Figs 2 and 1, respectively) were best preserved with the use of pinacidil compared with control hearts. Adenosine alone was not cardioprotective, and the addition of adenosine to pinacidil cardioplegia did not have any beneficial effect on postischemic recovery. Others have shown that adenosine possesses cardioprotective properties in crystalloid-perfused hearts, as an adjunct to hyperkalemia, or as a preconditioning agent [11, 14, 18]. However, no previous studies have demonstrated its effects as a single cardioplegic agent or PCO adjunct during global ischemia in a blood-perfused isolated heart model. Fundamental differences between crystalloid and blood perfusion, described subsequently, as well as widely varying physiologic protocols, preclude a clear comparison between this and previous studies.
There is excellent evidence that the A1 receptor is coupled to the KATP channel by a G-protein [25] and that the KATP channel represents the final common pathway for adenosine-mediated myocardial protection [26, 27]. However, it is possible that exogenous adenosine in the dosages used in this study did not exert a significant effect on KATP channels during global ischemia. This would be in agreement with previous data that have shown that endogenous adenosine does not play an important role in the activation of KATP channels [28]. Although it is clear that adenosine does not depolarize the cell membrane [24], this was not sufficient in this study to provide cardioprotection. Therefore, the avoidance of membrane depolarization alone may not be sufficient to fully explain PCO-mediated myocardial protection. Action potential shortening and the prevention of calcium overload during ischemia during PCO infusion may play more important roles. This is supported by the fact that adenosine does not cause action potential shortening in rabbit myocytes, as opposed to PCOs, which have been shown to significantly shorten ventricular action potential duration in the isolated rabbit heart [29, 30].
Induction of electromechanical arrest
Traditional cardioplegic solutions have been designed to induce rapid electromechanical arrest to minimize energy consumption during the period of global ischemia. Potassium-channel openers have typically shown prolonged times to electromechanical arrest compared with hyperkalemic cardioplegia [9, 10]. Both pinacidil and adenosine, used alone or in combination, resulted in shorter times to electrical and mechanical arrest than controls. However, pinacidil resulted in significantly more rapid mechanical arrest than adenosine (see Table 1). Potassium-channel openers mechanically arrest the heart by causing a dramatic shortening of the action potential with decreased calcium influx, which results in contractile failure [9]. Previous work has shown that there is no action potential shortening in ventricular myocardium during adenosine infusion [24]. This explains the slow mechanical arrest with this drug. The persistent energy consumption that results from the prolonged mechanical activity is one of the reasons that adenosine was not an effective cardioplegic agent at the doses examined in this study.
Postreperfusion arrhythmias
Potassium-channel openers have been shown to be proarrhythmic [9, 10, 31]. By decreasing action potential duration and therefore refractory period, these agents predispose to reentrant ventricular arrhythmias. This represents a potential drawback of this class of drugs.
Adenosine lacked this profibrillatory activity in our study. This is not a surprising finding in light of the fact that adenosine does not shorten action potential duration. This also suggests that adenosine is not effective in activating KATP channels in the ischemic heart in this model.
Coronary blood flow
Pinacidil is associated with a prolonged increase in reperfusion coronary flow when compared with the control group, which confirms previous results from our laboratory [9]. This was similarly seen in the pinacidil + adenosine group. This increase in flow persisted for the first 5 minutes of reperfusion before returning to baseline. There was no increase in postreperfusion flow in the adenosine-only group. The improvement in blood flow in the pinacidil group may be secondary to the better myocardial protection in these groups and the amelioration of the no-reflow phenomenon. The reasons for this hyperemic response also may be related to several other factors including the known vasodilator effect of pinacidil on vascular smooth muscle [32], improved endothelial protection [33], or a greater oxygen debt during ischemia [13]. Further studies are needed to clarify these issues.
Advantages and disadvantages of the blood-perfused isolated heart Langendorff model
The advantages and drawbacks of our model have been previously described [9, 10]. The use of a support animal can introduce variability related to ionic and hormonal fluctuations. However, continuous, careful monitoring of hemodynamic parameters and anesthesia with prompt correction of abnormalities in this study yielded a stable and reproducible preparation.
The more physiologic nature of this model is a distinct advantage over nonparabiotic and crystalloid-perfused models. The use of blood perfusion more closely approximates the clinical situation. The absence of hemoglobin limits the oxygen-carrying capacity of crystalloid solutions and alters coronary flow. Additionally, the absence of plasma proteins lowers oncotic pressures and, therefore, increases tissue edema when compared with blood perfusate [10]. Many of the protein moieties in blood, for example, histidine and erythrocyte carbonic anhydrase, are critical buffering elements that play a role in the organism’s response to ischemia. However, although this model offers a closer approximation to the clinical scenario than a crystalloid-perfused model, care should be taken in extrapolating results from in vitro studies to the clinical setting.
Summary
These experiments confirm our previous data that the PCO pinacidil is an effective cardioplegic agent. Pinacidil cardioplegia resulted in improved functional recovery compared with control and adenosine cardioplegia. No cardioprotective advantage was observed with the addition of adenosine to pinacidil. Adenosine alone was not an effective cardioplegic agent, despite shorter times to electrical and mechanical arrest than in control hearts. Our data suggest that adenosine does not play a significant role as a KATP channel opener in this model.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
1-Adrenoceptor activation increases ecto-5'-nucleotidase activity and adenosine release in rat cardiomyocytes by activating protein kinase C. Circulation 1995;91:2226-2234.This article has been cited by other articles:
|
|
 |
|
  G. P. Dobson and M. W. Jones Adenosine and lidocaine: a new concept in nondepolarizing surgical myocardial arrest, protection, and preservation J. Thorac. Cardiovasc. Surg., March 1, 2004; 127(3): 794 - 805. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
